Multiple Linear Regression Modeling of Nanosphere Self-Assembly via Spin Coating

Area of Science:

• Materials Science
• Nanotechnology
• Surface Science

Background:

• Nanosphere lithography utilizes self-assembled nanospheres for nanoscale patterning.
• Achieving large-area, defect-free nanosphere arrays is essential for advanced applications.
• Spin coating offers a high-throughput method for nanosphere self-assembly, but parameter optimization is challenging.

Purpose of the Study:

• To investigate the impact of seven process parameters on nanosphere self-assembly via spin coating.
• To identify statistically significant factors influencing hexagonal close packing and macroscale coverage.
• To develop predictive models for optimizing nanosphere lithography template fabrication.

Main Methods:

• A design-of-experiments approach was employed.
• Seven factors were systematically varied: nanosphere wt%, methanol/water ratio, solution volume, wetting time, spin time, maximum RPM, and ramp rate.
• Linear regression models were used to analyze two response variables: percentage hexagonal close packing and macroscale coverage.

Main Results:

• Statistical models identified significant main and two-way interaction effects on both response variables.
• Optimized settings were predicted based on the regression models.
• A trade-off was observed between ramp rate (for coverage) and shear/evaporation rates (for packing quality).

Conclusions:

• The study provides a data-driven approach to optimize nanosphere spin coating parameters.
• Understanding parameter interactions is crucial for fabricating high-quality nanosphere templates.
• Balancing competing factors is necessary to achieve both high coverage and ordered packing for nanosphere lithography.